System and method for follow-up local feature matching based on multiple functional-anatomical feature layers

Abstract

A method includes obtaining functional and anatomical image data sets from a subject acquired at different dates. The method includes receiving a volumetric coordinate of interest in a specified functional and anatomical image data set. The method includes generating a 3D feature matching map for at least one functional feature layer type and for at least one anatomical feature layer type for each non-specified functional and anatomical image data set relative to the specified functional and anatomical image data set utilizing the volumetric coordinate of interest. The method includes generating a best matching coordinate and a corresponding confidence metric value for each 3D feature matching map. The method includes calculating an optimal matching coordinate to the volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values and outputting a respective optimal matching coordinate for each of the non-specified functional and anatomical image data sets.

The invention claimed is: 
     
       1. A computer-implemented method for follow-up local feature matching, comprising:
 obtaining, via a processor, a plurality of functional and anatomical image data sets from a subject, each functional and anatomical image data set being acquired at a different date and comprising a functional image volume and a corresponding anatomical image volume; 
 receiving, via the processor, a first volumetric coordinate of interest in a specified functional and anatomical image data set from among the plurality of functional and anatomical image data sets; 
 generating, via the processor, a three-dimensional (3D) feature matching map for at least one functional feature layer type and for at least one anatomical feature layer type for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set utilizing the first volumetric coordinate of interest; 
 generating, via the processor, a best matching coordinate and a corresponding confidence metric value for each 3D feature matching map; 
 calculating, via the processor, a first optimal matching coordinate to the first volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set; and 
 outputting, via the processor, a respective first optimal matching coordinate for each of the non-specified functional and anatomical image data sets. 
 
     
     
       2. The method of  claim 1 , comprising:
 receiving, via the processor, a second volumetric coordinate of interest in the specified functional and anatomical image data set; 
 generating, via the processor, the 3D feature matching map for the at least one functional feature layer type and for the at least one anatomical feature layer type for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set utilizing the second volumetric coordinate of interest; 
 generating, via the processor, the best matching coordinate and the corresponding confidence metric value for each 3D feature matching map; 
 calculate, via the processor, a second optimal matching coordinate to the second volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set; and 
 output, via the processor, a respective second optimal matching coordinate for each of the non-specified functional and anatomical image data sets. 
 
     
     
       3. The method of  claim 1 , comprising determining the at least one functional feature layer type and the at least one anatomical feature layer type. 
     
     
       4. The method of  claim 1 , wherein the first volumetric coordinate of interest is in one of either the functional image volume or the corresponding anatomical image volume of the specified functional and anatomical image data set. 
     
     
       5. The method of  claim 1 , wherein the at least one functional feature layer type and the at least one anatomical feature layer type each comprises an entirety of image data or a portion of the image data above a threshold. 
     
     
       6. The method of  claim 1 , comprising displaying, via the processor, the first volumetric coordinate of interest with the specified functional and anatomical image data set and displaying both the first volumetric coordinate of interest and respective optimal matching coordinate with each non-specified functional and anatomical image data set on a display. 
     
     
       7. The method of  claim 1 , wherein calculating the first optimal matching coordinate is based on the best matching coordinates and their corresponding confidence metric values of all generated 3D feature matching maps for a respective non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set. 
     
     
       8. The method of  claim 1 , wherein generating the 3D feature matching map for the at least one functional feature layer type and for the at least one anatomical feature layer type comprises:
 determining search parameters; 
 determining a feature matching metric to search; 
 extracting a fixed sub-volume from a first image volume centered on the first volumetric coordinate of interest, wherein the first image volume is from the specified functional and anatomical image data set; 
 extracting a plurality of sub-volumes at different positions from a second image volume based on the search parameters, wherein the second image volume is from one of the unspecified functional and anatomical image data set; 
 calculating a respective matching metric between the fixed sub-volume and each respective sub-volume of the plurality of sub-volumes and assigning the respective matching metric to a corresponding position in the 3D feature matching map; and 
 outputting the 3D feature matching map. 
 
     
     
       9. The method of  claim 8 , wherein the first image volume and the second image volume both comprise either the at least one functional feature layer or the at least one anatomical feature layer type. 
     
     
       10. A system for follow-up local feature matching, comprising:
 a memory encoding processor-executable routines; 
 a processor configured to access the memory and to execute the processor-executable routines, wherein the routines, when executed by the processor, cause the processor to: 
 obtain a plurality of functional and anatomical image data sets from a subject, each functional and anatomical image data set being acquired at a different date and comprising a functional image volume and a corresponding anatomical image volume; 
 receive a first volumetric coordinate of interest in a specified functional and anatomical image data set from among the plurality of functional and anatomical image data sets; 
 generate a three-dimensional (3D) feature matching map for at least one functional feature layer type and for at least one anatomical feature layer type for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set utilizing the first volumetric coordinate of interest; 
 generate a best matching coordinate and a corresponding confidence metric value for each 3D feature matching map; 
 calculate a first optimal matching coordinate to the first volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set; and 
 output a respective first optimal matching coordinate for each of the non-specified functional and anatomical image data sets. 
 
     
     
       11. The system of  claim 10 , wherein the routines, when executed by the processor, cause the processor to:
 receive a second volumetric coordinate of interest in the specified functional and anatomical image data set; 
 generate the 3D feature matching map for the at least one functional feature layer type and for the at least one anatomical feature layer type for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set utilizing the second volumetric coordinate of interest; 
 generate the best matching coordinate and the corresponding confidence metric value for each 3D feature matching map; 
 calculate a second optimal matching coordinate to the second volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set; and 
 output a respective second optimal matching coordinate for each of the non-specified functional and anatomical image data sets. 
 
     
     
       12. The system of  claim 10 , wherein the first volumetric coordinate of interest is in one of either the functional image volume or the corresponding anatomical image volume of the specified functional and anatomical image data set. 
     
     
       13. The system of  claim 10 , wherein the routines, when executed by the processor, cause the processor to display the first volumetric coordinate of interest with the specified functional and anatomical image data set and display both the first volumetric coordinate of interest and respective optimal matching coordinate with each non-specified functional and anatomical image data set on a display. 
     
     
       14. The system of  claim 10 , wherein calculating the first optimal matching coordinate is based on the best matching coordinates and their corresponding confidence metric values of all generated 3D feature matching maps for a respective non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set. 
     
     
       15. The system of  claim 10 , wherein generating the 3D feature matching map for the at least one functional feature layer type and for the at least one anatomical feature layer type comprises:
 determining search parameters; 
 determining a feature matching metric to search; 
 extracting a fixed sub-volume from a first image volume centered on the first volumetric coordinate of interest, wherein the first image volume is from the specified functional and anatomical image data set; 
 extracting a plurality of sub-volumes at different positions from a second image volume based on the search parameters, wherein the second image volume is from one of the unspecified functional and anatomical image data set; 
 calculating a respective matching metric between the fixed sub-volume and each respective sub-volume of the plurality of sub-volumes and assigning the respective matching metric to a corresponding position in the 3D feature matching map; and 
 outputting the 3D feature matching map. 
 
     
     
       16. The system of  claim 15 , wherein the first image volume and the second image volume both comprise either the at least one functional feature layer or the at least one anatomical feature layer type. 
     
     
       17. A non-transitory computer-readable medium, the computer-readable medium comprising processor-executable code that when executed by a processor, causes the processor to:
 obtain a plurality of functional and anatomical image data sets from a subject, each functional and anatomical image data set being acquired at a different date and comprising a functional image volume and a corresponding anatomical image volume; 
 receive a first volumetric coordinate of interest in a specified functional and anatomical image data set from among the plurality of functional and anatomical image data sets; 
 generate a three-dimensional (3D) feature matching map for at least one functional feature layer type and for at least one anatomical feature layer type for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set utilizing the first volumetric coordinate of interest; 
 generate a best matching coordinate and a corresponding confidence metric value for each 3D feature matching map; 
 calculate a first optimal matching coordinate to the first volumetric coordinate of interest based on the best matching coordinates and their corresponding confidence metric values for each non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set; and 
 output a respective first optimal matching coordinate for each of the non-specified functional and anatomical image data sets. 
 
     
     
       18. The non-transitory computer-readable medium of  claim 17 , wherein the first volumetric coordinate of interest is in one of either the functional image volume or the corresponding anatomical image volume of the specified functional and anatomical image data set. 
     
     
       19. The non-transitory computer-readable medium of  claim 17 , wherein calculating the first optimal matching coordinate is based on the best matching coordinates and their corresponding confidence metric values of all generated 3D feature matching maps for a respective non-specified functional and anatomical image data set of the plurality of functional and anatomical image data sets relative to the specified functional and anatomical image data set. 
     
     
       20. The non-transitory computer-readable medium of  claim 17 , wherein the processor-executable code, when executed by the processor, cause the processor to display the first volumetric coordinate of interest with the specified functional and anatomical image data set and display both the first volumetric coordinate of interest and respective optimal matching coordinate with each non-specified functional and anatomical image data set on a display.